Two methods of intravenously administering radionuclides are the bolus and the continously decreasing infusion (to provide a plateau nf radioactivity in arterial blood throughout the scanning period). We propose to determine which method provides scan data that is more indicative of myocardial perfusion in man. Evaluation will be by serial scanning (5.5 minutes/scan) which will be started simultaneously with the bolus or six minutes into the continuously decreasing infusion. Multiple radionuclides (K43, Rb81, Rb82, Th201) will be used to minimize the time interval between comparison studies. The psotron emissions of Rb81 will be utilized in determining whether the better spatial resolution (with decreased ount rates) of coincidence counting is necessary for myocardial mapping. Data handling has been computerized for averaging scan elements, summing scan elements and for computing rate of uptake and appearance time values. These mathematical methods will also be assessed to determine the combination of computations and display techniques (including intensity mcdulated colors) that will provide the clinician with the most meaningful data in the shortest possible time.